The invention relates to apparatus for mixing and subsequently separating immiscible or semi-immiscible fluids, particularly liquid solvents, for transferring a solute from one fluid to the other. The three liquids are termed the light, intermediate and a third phase variously termed the lower, residual, heavy or stationary phase.
Techniques of extraction for separation of materials have been the subject of much experimentation. It is well established that conventional funneling techniques will not, on a single extraction pass, suffice for the separation of similar substances. Within this technology, a technique, "countercurrent distribution" (CCD), has been developed to utilize small differences in partitioning properties for purification purposes by repetitive contacting extraction passes. A hallmark of CCD is that the distribution of solute in the train can be mathematically predicted, thereby providing a model for defining the purity of the isolated product. The theory is set forth in King and Craig, "Countercurrent Distribution," Methods of Biochemical Analysis, Volume 10, page 201.
Devices built for CCD usually consist of a series of interconnected glass tubes so configured that in each tube an upper (light) phase can be mixed with a lower (residual) phase, allowed to separate from the lower phase and then automatically transferred to the next tube in the series while leaving the lower phase behind. These devices operating on a stepwise transfer basis allow precise fractionation of complex mixtures even in cases where only small amounts of material are available.
As indicated in a CCD train, only one phase, usually the upper phase, is transferred, and the other remains stationary in each element of the train. An important modification of CCD has been developed in which both phases move at a predicated and controlled rate through the train. This is known as "counter-double-current distribution" (CDCD). This technique is important because it allows a fresh portion of the sample to be introduced at a central point in the array on each transfer. CDCD mathematical distribution theory has been reported in Post and Craig "A New Type of Stepwise Countercurrent Distribution Train," 35 Anal. Chem. 641 (1963). In CDCD, two distribution techniques are available: (1) upper stage transfer in one direction on every other equilibration stage while lower stage transfer takes place in an opposite direction on intervening transfers; and (2) both stages are transferred to an adjacent tube, in opposite directions on each equilibration.
Within the prior art, a variety of such devices have been reported which are used for CCD or CDCD solvent phase extraction. Typical are U.S. Pat. Nos. 2,892,688, 2,895,808 and 2,973,250. Additionally, the technical literature is replete with reports of experimental devices used in a CCD mode. Typical are Gaucher, "An Introduction to Chromatography," Volume 46, J. Chem. Ed. 729 (1969); Alderweireldt, "New Instrument for Continued Batch Wise Separation by Extraction," Analytical Chemistry 30, 1290 (1961); and Hietala, "A Counter-Current Distribution Method," Ann. Acad. Scient. Fennicae, A.I.I. 100 at page 13.
More pertinent prior art appears in the Post and Craig article, which shows an apparatus utilizing the technique practicing the first technique of CCD set forth above using side-arm decantation of the upper phase. This technique, however, leads to co-flow. This type of system has not achieved wide commercial success because the tubes of the apparatus are individually constructed, which require a large number of joints. By having so many joints between system components, the propensity for leakage is increased. Also, the device of Post and Craig requires operation in two directions, therefore, necessitating a complicated drive train.
A device which has achieved a commercial success in the practice of CCD is described by Raymond, "Compact Counter Current Distribution Apparatus,": Analytical Chem. 30, 1214 (1958). This device is also explained in detail in U.S. Pat. No. Re. 25,186 (Re. 25,186) to Raymond.
The Raymond device shows best in FIGS. 1 and 2 a system which is mounted for rotation about bearing 19 having a number of mixing and separating elements 14. The elements are constructed as common elements individually shown in FIGS. 2 and 5. Each of the elements contains a mixing phase tube 34 and a phase separating tube 35. The tubes are coupled by a common bifurcated juncture 36 shown best in FIG. 3. Shown in FIG. 3, the mixing tube has a closed end 37 with a cap 38, and an effluent phase entrance tube couples the tube 34 to the phase separating tube 35 via down spout 51.
In operation of the Raymond device, each of the elements 14 are filled with a predetermined amount of lower (residual) phase material equal to the volume which can be retained between the down spout element 51 and the lower septum wall 50. This is generally known as the critical volume. A quantity of the effluent or upper phase is introduced generally, while the elements are in the horizontal position with the reservoir 49 disposed upwardly. The quantity of the effluent phase is not critical but limited only to the capacity of the mixing tube 34. Operation commences with the phase separating tube 35 disposed above the mixing tube 34 with the elements 14 rocked to thoroughly mix the effluent and residual phases. The elements are then rotated 180.degree. into the position shown in FIG. 4 of the Raymond device such that the intermixed contents are allowed to flow into the separating tube 35. Because of the relatively large area of the separating phase tube, separation occurs, and a quantity of the effluent phase may enter the down spout 51. However, none of the critical residual phase will be drained.
The 90.degree. rotation then takes place with the residual phase trapped in the reservoir 49, as shown in the Raymond patent, and the separated effluent phase will pass through the down spout 51 into the drain tube 53 with this action. The effluent phase is deposited in the next adjacent element 14 so that on the next rotational cycle the second extraction can be performed.
In the above-mentioned U.S. Pat. No. 2,892,688 to Buchler there is disclosed a mixing and separating device in which the tubes are interconnected in a cascade arrangement. It does not appear possible to arrange the tubes in a parallel bank. Moreover, in this arrangement, it does not appear possible to retain either of the mobile phases for mixing to carry out conventional countercurrent fractionation with two stationary phases and one mobile phase. A similar device is disclosed by Meltzer, "Three-phase Counter Current Distribution: Theory And Application To The Study of Strandin", Anal. Chem. 30, 1290 (1961).
Similarly, the device disclosed by Wilhelm and Foos, "A Counter Current Liquid-Liquid Extractor", Ind. Eng. Chem. 51, 633 (1959), does not appear to have the capability of operating with three phases. Moreover, co-current flow takes place in their apparatus. With reference to FIG. 2 of the referenced article, placement of tube H as they have indicated necessitates co-current flow of the lighter phase with the heavier phase because in the sequence of rotation the lighter phase will contact tube H and some of it will be trapped above the heavier phase as the rotation continues until decantation by tube L. On continued rotation, the heavier phase will push the small amount of lighter phase trapped in tube H into the mixer-settler stage M'. The Wilhelm and Foos device is also similar to that of Alderweireldt as described in his article referenced above.